AI RESEARCH
All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG
arXiv CS.CL
•
ArXi:2604.20199v1 Announce Type: new Multilingual Retrieval-Augmented Generation (mRAG) leverages cross-lingual evidence to ground Large Language Models (LLMs) in global knowledge. However, we show that current mRAG systems suffer from a language bias during reranking, systematically favoring English and the query's native language. By